Extract the fitted linear model for a PrestoGP model

This method return the fitted linear model (of class cv.glmnet or cv.ncvreg) for a PrestoGP model.

Usage

# S4 method for class 'PrestoGPModel'
get_linear_model(model)

Arguments

model: The PrestoGP model object

Value

The fitted linear model (of class cv.glmnet or cv.ncvreg).

Details

It is important to note that the model is fit to the transformed data. The CV error rate and predicted values of Y will not be correct for the original (untransformed) data. This method should be used primarily for examining the coefficient paths and generating plots.

References

Messier, K.P. and Katzfuss, M. "Scalable penalized spatiotemporal land-use regression for ground-level nitrogen dioxide", The Annals of Applied Statistics (2021) 15(2):688-710.

Examples

data(soil)
soil <- soil[!is.na(soil[,5]),] # remove rows with NA's
y <- soil[,4]                   # predict moisture content
X <- as.matrix(soil[,5:9])
locs <- as.matrix(soil[,1:2])

soil.vm <- new("VecchiaModel", n_neighbors = 10)
soil.vm <- prestogp_fit(soil.vm, y, X, locs)
#> 
#> Estimating initial beta... 
#> Estimation of initial beta complete 
#> 
#> Beginning iteration 1 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 1 complete 
#> Current penalized negative log likelihood: 487.7521 
#> Current MSE: 9.104869 
#> Beginning iteration 2 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 2 complete 
#> Current penalized negative log likelihood: 483.0906 
#> Current MSE: 9.024679 
#> Beginning iteration 3 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 3 complete 
#> Current penalized negative log likelihood: 482.6058 
#> Current MSE: 9.06609 
#> Beginning iteration 4 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 4 complete 
#> Current penalized negative log likelihood: 482.4715 
#> Current MSE: 9.05603 
#> Beginning iteration 5 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 5 complete 
#> Current penalized negative log likelihood: 482.3219 
#> Current MSE: 9.050498 
#> Beginning iteration 6 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 6 complete 
#> Current penalized negative log likelihood: 482.2806 
#> Current MSE: 9.049297 
#> Beginning iteration 7 
#> Estimating theta... 
#> Estimation of theta complete 
#> Estimating beta... 
#> Estimation of beta complete 
#> Iteration 7 complete 
#> Current penalized negative log likelihood: 482.2806 
#> Current MSE: 9.054167 
get_linear_model(soil.vm)
#> 
#> Call:  cv.glmnet(x = as.matrix(model@X_tilde), y = as.matrix(model@y_tilde),      nfolds = nfolds, foldid = foldid, parallel = parallel, relax = penalty ==          "relaxed", alpha = model@alpha, family = family, penalty.factor = pen.factor) 
#> 
#> Measure: Mean-Squared Error 
#> 
#>      Lambda Index Measure      SE Nonzero
#> min 0.00681    27  0.5276 0.06112       4
#> 1se 0.07646     1  0.5408 0.06193       0